Note where the AD tools are going: DifferentialEquations - Derivatives in ODE function/ nesting AD - #2 by ChrisRackauckas . I think most of these issues should be handled by what we’re moving things towards.
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
Neural Networks combined with Diffeq | 26 | 1321 | May 7, 2021 | |
Tracking initial condition to optimize starting value too Diffeqflux | 11 | 1298 | July 25, 2019 | |
DifferentialEquations.jl + Flux.jl or Knet.jl | 15 | 3594 | October 18, 2018 | |
Different loss function for training ODE or DDE? | 5 | 569 | January 20, 2021 | |
DiffEqFlux: SOSRA2() algorithm + BacksolveAdjoint() gives an error | 2 | 541 | May 23, 2021 |